Somebody sold you a content calendar. Two posts a week, consistency wins, the compound effect. So you started publishing, and eighteen months later you have 140 blog posts and less organic traffic than when you had 12.
This is common enough that it deserves a name. Publishing more content is not neutral. Done badly it makes a site measurably worse, in four specific ways.
One: your pages start competing with each other
You write "Best Time to Trek in Nepal". Four months later, a different writer produces "When to Visit Nepal for Trekking". Then "Nepal Trekking Seasons Explained".
Three pages, one query. Google has to choose. It splits signals between them, sometimes ranks the weakest, and sometimes rotates between them week to week. Any links you earn are divided across three URLs instead of concentrated on one.
This is keyword cannibalisation and it is almost universal on sites that publish to a schedule, because the schedule creates pressure to produce topics rather than to check whether the topic already exists.
You can find it in one search: site:yourdomain.com trekking seasons. If three of your own pages come back, you have found it.
Two: the average drags the whole site down
Google assesses sites, not only pages. A domain where most content is thin gets treated as a domain that produces thin content, and that assessment affects the good pages too.
So a site with 12 strong pages can outrank a site with 12 strong pages and 128 weak ones. The 128 are not inert. They are evidence.
This is the part people find hardest to accept, because the weak pages each get a trickle of traffic and deleting them feels like throwing away results. The trickle is not worth what it costs.
Three: crawl budget goes to the wrong places
Googlebot does not crawl everything on your site every day. It allocates. On a 140-page site where 128 pages have never earned a link or a click, a meaningful share of that allocation gets spent re-checking pages that will never rank.
The visible symptom is that new pages take a long time to appear in search, and updates to important pages take a long time to register. Owners usually blame Google. It is usually a self-inflicted signal-to-noise problem.
Four: a schedule and quality pull in opposite directions
This is the underlying cause of the other three.
When the commitment is two posts a week, the deadline is the fixed thing and quality becomes the variable. Nobody decides to write something thin. It happens because Thursday arrives.
The genuinely useful things — original data, a real example, a position someone might disagree with — take longer than the slot allows. So they get cut, week after week, and what remains is the part that can be produced to deadline: a restatement of what already ranks.
That was always weak. It is now close to worthless, because an AI answer can restate the consensus better and faster than your blog can, and a page that only restates existing sources cannot be cited over them. I have written more about that in Google Search vs AI Search.
How to find out if this is you
Open Search Console, Pages report, last six months, sorted by clicks. Then read the bottom of the list.
Count the pages with zero clicks in six months. Then count the pages with fewer than five.
If more than half your pages are in that second group, this article is about your site. That is not a disaster and it is very fixable — but adding more pages will make it worse, not better.
Pruning: what to do with each page
Go through the low performers and sort them into four piles. Be honest and do it quickly; second-guessing each one is how this project dies.
Merge. Several pages on the same topic become one genuinely comprehensive page. Take the best sections of each. Keep the strongest URL — usually the one with links or the most history — and 301 redirect the others to it. This is the highest-value action on the list, because it converts three weak signals into one strong one.
Improve. The page targets a query that matters and does it badly. Rewrite it properly. Usually a small number of pages, and worth real effort.
Delete and redirect. No traffic, no links, no strategic purpose, and nothing worth merging. Redirect it to the most relevant page that still exists. Do not redirect everything to the homepage — that is a soft 404 and Google treats it as one.
Leave it. Some pages exist for customers rather than search: terms, policies, an announcement from 2024. They are not competing for anything. Leave them alone and stop counting them.
Pruning does not always raise traffic, and I would be overselling it to promise that. What it reliably does is concentrate your signals, make your reporting legible, and stop new work being buried.
What to do instead of a calendar
Replace "two posts a week" with a fixed number of pages you intend to be the best available answer for. Eight is a reasonable target for a small business.
Choose them by intent, not volume — the queries with a customer behind them, which I have written about in the piece on search intent. Then spend the time you would have spent on posts nine through eighty making those eight genuinely better: real prices, real examples, real photographs, answers to the questions people actually ask before buying.
Update them when something changes. A page you revisit four times over two years will outperform four separate pages, and it is less work.
The part that is hard
This is a difficult recommendation to accept because it looks like doing less, and because somebody is usually being paid for volume. A report saying "published eight articles" reads better than one saying "improved three pages and deleted forty".
The second report is the one describing a site that is getting better.
If you have a large site and want a view on what to merge, improve or delete, send me the Search Console export. Sorting it is usually a single afternoon, and the list it produces tends to be shorter and more obvious than people expect.